← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
8k
tokens
835.3
PP tok/s
72.6
TG tok/s
9807
TTFT (ms)
20.4
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.2.12
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 894.2 | 79.1 | 20.1 GB | view |
| 4k | 909.6 | 76.9 | 20.2 GB | view |
| 8k | 835.3 | 72.6 | 20.4 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 79.1 | 1.00× |
| 2× | 128.4 | 1.62× |
| 4× | 163.7 | 2.07× |
| 8× | 197.4 | 2.50× |